Skip to content

A python program that implements Aspect Based Sentiment Analysis classification system for SemEval 2016 Dataset.

Notifications You must be signed in to change notification settings

thestrox/Aspect-Based-Sentiment-Analysis

Repository files navigation

Aspect-Based-Sentiment-Analysis

Aspect Based Sentiment Analysis task focuses on the recognition of aspect category and classification of emotions (positive, negative, neutral) in text. The aim of Aspect Based Sentiment Analysis is to determine the sentiment polarity expressed towards certain aspect. This system is based on supervised learning using Support Vector Machine (SVM). Multi label svc classifier is used to classify the laptop reviews into 20 aspect categories and also to classify the review as positive, negative or neutral. For each sentiment, a multi label svc classifier is used. Performance of the system is also compared with other classifiers.

Dataset

Dataset is avaibale at SemEval website.

Required libraries

  • Scikit Learn
  • Nltk
  • pandas
  • numpy

P.S. This is a Mini-Project for a 2 month crash course (NLP) in the Final Year at the Department of Computer Science and Engineering, Kathmandu University.

Releases

No releases published

Packages

No packages published